Re: Hofman and Diaby talk about P=NP at INFORMS 2007



On Feb 13, 10:37 am, t...@xxxxxxxxxxxxx wrote:
In article <1171376630.989060.181...@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>,

<moustapha.di...@xxxxxxxxxxxxxxxxxx> wrote:
There is no *mathematical* connection between the polytope of my model
and the TSP polytope. Proposition 1 in my paper shows equivalence of
the feasible set of my model and the assignment polytope. You get a
TSP solution out of it only by interpretation. ...The connection to a
TSP solution is only cognitive (The mathematical connection is to the
assignment polytope as proved in Proposition 1.).

Of course you didn't build your model with what you call a "mathematical"
correspondence to the TSP polytope in mind, and the connection remains
only "cognitive" in *your* mind. However, once your model has been
constructed, one can study it and derive further consequences from
its existence. This is what Moews has done, and he has shown that
from your model, one can derive a *mathematical* connection to the TSP
polytope---


And, which exact formulation of the the TSP polytope did he use to da
that again? ...You don't know what you are saying.

//MD

.